Abstract
6-18F-Fluorodopamine (18F-FDA) PET is a highly sensitive tool for the localization of pheochromocytoma (PHEO). The aim of this study was to establish cutoff values for pathologic and physiologic adrenal gland tracer uptake. Methods: 18F-FDA PET with CT coregistration was performed in 14 patients (10 men and 4 women; age [mean ± SD], 42.9 ± 13.3 y) with unilateral adrenal gland PHEO and in 13 control subjects (5 men and 8 women; age, 51.7 ± 12.5 y) without PHEO. Standardized uptake values (SUVs) were compared between adrenal glands with PHEO and normal left adrenal glands in control subjects. Results: 18F-FDA accumulation was observed in all adrenal glands with PHEO and in 6 of 13 control adrenal glands (P = 0.02). The SUV was higher in adrenal glands with PHEO (mean ± SD, 16.1 ± 6.1) than in 18F-FDA–positive control adrenal glands (7.7 ± 1.4) (P = 0.005). SUV cutoffs for distinguishing between adrenal glands with PHEO and normal adrenal glands were 7.3 (100% sensitivity) and 10.1 (100% specificity). Conclusion: The SUVs of adrenal foci on 18F-FDA PET facilitate the distinction between adrenal glands with PHEO and normal adrenal glands.
Pheochromocytomas (PHEOs) are rare catecholamine-producing tumors of the adrenal medulla (1,2). The diagnosis of PHEO can be reliably confirmed or excluded by the biochemical parameter of catecholamine excess, in particular, fractionated metanephrines in urine or plasma (3). The localization of PHEO usually requires both anatomic and functional imaging studies. Agents that specifically target the catecholamine storage and secretion pathways include 123I/131I-metaiodobenzylguanidine and 6-18F-fluorodopamine (18F-FDA) (4,5). In our experience, 18F-FDA PET is a highly sensitive tool for localizing PHEO (4,6) but may lead to false-positive results because of physiologic uptake by normal adrenal glands. This uptake can be particularly misleading in patients who are prone to bilateral PHEOs because of underlying genetic abnormalities (7,8).
The aim of this study was to establish cutoff values for pathologic and physiologic adrenal gland tracer uptake for 18F-FDA PET. For this purpose, the distributions of 18F-FDA in adrenal glands and other tissues in patients with benign PHEO and control subjects without PHEO were compared.
MATERIALS AND METHODS
Patients
A total of 104 consecutive patients were referred for known or suspected PHEO and underwent 18F-FDA PET/CT between March 2005 and June 2006. After exclusion of patients with extraadrenal or metastatic PHEO, 14 patients (10 men and 4 women; age [mean ± SD], 42.9 ± 13.3 y) with histologically confirmed adrenal gland PHEO were studied. Underlying genotypes are indicated in Table 1. The control group consisted of 13 subjects (5 men and 8 women; age, 51.7 ± 12.5 y). Reasons for PHEO evaluation are indicated in Table 1. In all control subjects, PHEO was ruled out by normal plasma free metanephrine levels and clinical follow-up (3). The study protocol was approved by the Institutional Review Board of the National Institutes of Child Health and Development, National Institutes of Health. All patients provided written informed consent.
Adrenal Imaging Results
18F-FDA PET
18F-FDA PET was performed as previously described (9) with a Discovery ST PET/CT scanner (GE Healthcare). The injected 18F-FDA dose was typically 37 MBq, that is, a mean mass of 400 (range, 180–810) μg of 18F-FDA with a mean specific activity of ∼38 (range, about 30–49) GBq/mmol.
Analysis of Data
18F-FDA PET/CT studies were read by a nuclear medicine physician who was unaware of the results of other investigations. Any visible adrenal foci of uptake higher than the background were considered 18F-FDA positive. Standardized uptake values (SUVs) corrected for lean body mass were calculated (SUV = [Bq/g per Bq injected] × lean body mass) with software from MedImage. Maximum SUVs were determined in manually drawn regions of interest over adrenal lesions and in normal left adrenal glands, as delineated by CT. The SUV was not calculated for 18F-FDA–negative right adrenal glands to exclude interference from physiologic uptake by the liver and the biliary tract. Left adrenal gland SUVs of control subjects served as a reference for normal uptake.
Regions of interest were also drawn manually around the parotid gland, the thyroid gland, and the myocardium and in 4 consecutive slices around the right lung (level below the carina), the liver (upper part), the spleen (middle part), and the pancreas (body), as delineated by CT. Under the assumption of a homogeneous tracer distribution in the large organs, average SUVs were calculated for the central parts of the lungs, liver, and spleen. For the smaller organs, maximum SUVs were used, because estimates of average SUVs in smaller structures may be hampered by resolution limitations.
Statistics
Results are reported as mean ± SD. Fisher exact and unpaired Student t tests were used for comparisons of the numbers of 18F-FDA–positive adrenal glands and SUVs, respectively. A 2-sided P value of <0.05 was considered significant. A receiver operating characteristic (ROC) curve was constructed for different upper reference limits of SUVs (10). Statistical analysis was performed with the Statistical Package for the Social Sciences (SPSS for Windows 12; SPSS Inc.).
RESULTS
Uptake of 18F-FDA by Adrenal Glands with PHEO Versus Normal Adrenal Glands
CT or MRI showed an adrenal tumor in all patients with PHEO. All 14 tumors were 18F-FDA positive (Table 1 and Fig. 1). Six of 13 control left adrenal glands (46%) were 18F-FDA positive; 2 of these contained an incidentaloma (P = 0.02 for controls vs. adrenal glands with PHEO) (Table 1). Maximum SUVs were higher in adrenal glands with PHEO (16.1 ± 6.1; range, 7.3–26.3) than in 18F-FDA–positive control left adrenal glands (7.7 ± 1.4; range, 5.7–9.5) (P = 0.005) (Fig. 2). The mean SUV in all control left adrenal glands, that is,18F-FDA positive and negative, was 6.6 ± 2.0 (range, 3.8–9.8) (P < 0.001 for controls vs. adrenal glands with PHEO).
Cross-sectional 18F-FDA PET/CT images of patient with left adrenal gland PHEO (P10: 18F-FDA positive, SUV = 20.1) and 2 control subjects with normal left adrenal glands (C12: 18F-FDA positive, SUV = 9.5; C2: 18F-FDA negative, SUV = 3.9). Arrows indicate left adrenal gland areas.
18F-FDA PET SUVs in normal left adrenal glands and left or right adrenal glands with PHEO. SUVmax = maximum SUV.
The area under the ROC curve for adrenal gland SUVs was 0.962 (Fig. 3). To provide 100% sensitivity, the upper reference for a normal SUV was established at 7.3, resulting in a specificity of 69%. To provide 100% specificity, the upper reference for normal was established at 10.1, resulting in a sensitivity of 86%. With this cutoff value, false-negative results were obtained in 2 patients (patients 1 and 2). The diameters of these tumors were 3 cm (patient 1) and 3.4 cm (patient 2); the latter was hemorrhagic.
ROC curve for 18F-FDA PET SUVs.
Extraadrenal Tissue Distributions of 18F-FDA
Physiologic 18F-FDA uptake was observed in the salivary and thyroid glands, heart, lungs, liver, kidneys, pancreas, and bowel (Table 2 and Fig. 4), and values were similar between patients with PHEO and control subjects (Table 2).
Tissue distributions of 18F-FDA. Anterior reprojected 18F-FDA PET images of patient with left adrenal gland PHEO (P11) and control subject (C4). 1 = liver; 2 = spleen; 3 = lungs; 4 = parotid glands; 5 = thyroid gland; 6 = heart; 7 = kidneys; 8 = PHEO. Adrenal glands in control subject were considered 18F-FDA negative.
Organ Distributions of Physiologic 18F-FDA Uptake
DISCUSSION
18F-FDA PET is a promising tool for the localization of PHEO (4,11). 18F-FDA is actively transported into neurosecretory granules of catecholamine-producing cells via vesicular monoamine transporters after uptake into cells by the norepinephrine transporter (12,13). We previously observed excellent sensitivity of 18F-FDA PET for the localization of benign PHEO (4). However, physiologic 18F-FDA uptake by normal adrenal glands and extraadrenal tissues is a possible confounder in the identification of adrenal gland PHEO and extraadrenal paraganglioma or metastases, respectively. In the present study, almost half of the control subjects had 18F-FDA–positive results. In our previous study, only 2 of 11 PHEO-negative subjects had false-positive 18F-FDA PET results. However, systematic investigation of the adrenal glands in our previous study and other early 18F-FDA PET studies (6,14) was hampered by the lack of coregistered CT. Physiologic uptake is probably even higher when higher tracer doses are administered. In a pilot study (data not shown), we performed non–CT-coregistered 18F-FDA PET with a dose of 148 MBq instead of 37 MBq in 7 control subjects without PHEO. With this higher dose, all normal left adrenal glands except for one were 18F-FDA positive, with a mean SUV of 8.8 ± 3.2.
We have found that SUVs of adrenal foci on 18F-FDA PET can help distinguish between PHEO-related uptake and physiologic uptake in the adrenal glands. SUV cutoffs were established at <7.3 for physiologic uptake and >10.1 for PHEO-related uptake (100% specificity and 100% sensitivity, respectively). However, findings on 18F-FDA PET should always be interpreted in conjunction with other radiologic, biochemical, and clinical characteristics. In particular, patients with underlying gene mutations predisposing them to recurrent and bilateral PHEOs warrant careful follow-up despite negative PET results. Furthermore, in the diagnostic work-up of PHEO,18F-FDA PET is best positioned as a localizing tool, not as a screening method for the presence of a PHEO. As pointed out earlier, biochemical screening represents the gold standard for confirming or ruling out the diagnosis. Also, with respect to screening for bilateral PHEOs, it is unknown whether adrenal gland uptake of 18F-FDA is altered by previous surgical resection of the contralateral gland. Theoretically, compensatory hypertrophy of the remaining adrenal medulla could lead to enhanced 18F-FDA uptake and false-positive results.
CONCLUSION
In conclusion, calculation of SUVs of adrenal foci on 18F-FDA PET facilitates the distinction between PHEO-related and physiologic tracer accumulation in the adrenal glands. The diagnosis of PHEO is highly unlikely when the adrenal gland SUV is below 7.3 and very likely when it exceeds 10.1.
Acknowledgments
This research was supported by the Intramural Research Program of the NICHD, NIH, and, in part, by the Intramural Research Program of the Center for Cancer Research, National Cancer Institute, NIH. We thank Jacques Lenders for helpful discussions and feedback on the article.
Footnotes
-
COPYRIGHT © 2007 by the Society of Nuclear Medicine, Inc.
References
- Received for publication May 2, 2007.
- Accepted for publication August 24, 2007.
